8 published studies recovered

Measurement
you can verify.

You study whether interventions work.We make the outcome measure auditable.

Minutes
Question to launch
Not semesters
Zero
Vendor trust required
Verify it yourself
Free
For research use
No seat licenses
Open
Instrumentation
Audit any claim

How it works

Three phases. Full audit trail.

Every study follows the same pipeline — from design to independently verifiable results.

01

Design

Power calculator, pre-registration, primary outcome freeze. Server-enforced — no post-hoc switching.

02

Measure

Scaffold-aware proficiency tracking. Knows the difference between an unaided answer and a helped one.

03

Verify

Hash-chained records, standalone verifier, deterministic replay. Any third party can audit.

Capabilities

Four engines. One discipline.

D

Depth Engine

Proficiency measurement that distinguishes helped from unaided performance. Certifies mastery only when evidence survives scrutiny.

C

Calibrate

Item calibration with verifiable certificates. Fit diagnostics with root-cause analysis, not folk cutoffs.

A

Audit Harness

Known-truth validation. Runs your engine against synthetic learners alongside baselines — reports where it wins and where it loses.

W

Wrightyard

Content validation against domain rules. Every rule teaches: names the failing element and says how to fix it.

Validated

8 published studies recovered.

Synthetic data calibrated to published parameters. The platform recovers the known effect in every case.

VanLehn 2011

d = 1.003

Mueller & Dweck

d = 2.20

Hattie 2009

ρ = 0.83

Chi 2009

27.4%

Kraft 2021

N valid

See all 8 validations →

Verified on real data

Two public datasets. Every number recomputes.

Synthetic recoveries prove the engine finds known truths. These calibrations go further: real public data, novel numbers, and a certificate that lets you rerun everything yourself.

CalibrationRev · WO-HEFF-04

Finding Percents

ASSISTments skill builders, 2012–13 public release

212items
16,005responses
1.66separation
T1T2T3T4T5
b3e1a70a e5e2a4a9 … f2dc9b9⧉ copy
calibrate-0.1.1 · seed 20260824Verify this run →
CalibrationRev · WO-HEFF-04

Ordering Fractions

ASSISTments skill builders, 2012–13 public release

71items
7,624responses
1.32separation
T1T2T3T4T5
c8d6dfeb 2121208c … 96117eb⧉ copy
calibrate-0.1.1 · seed 20260824Verify this run →

Research

Read the papers.

Open research behind the platform. All published under open licenses.

Manifesto

The Open Measurement Layer

Why open education AI needs one more layer — and what it's made of.

Kumar Srivastava · July 2026 · CC-BY-4.0 · Coming soon

Paper 1

Embedded Measurement in Adaptive Learning

Scaffold-conditioned proficiency estimation with verifiable evidence trails.

EdArXiv Preprint · 2026 · Coming soon

Paper 2

The Open Measurement Layer v2

Architecture, validation, and the case for separating measurement from instruction.

EdArXiv Preprint · 2026 · Coming soon

Start measuring
what matters.

Free for research use. No dark patterns. Pseudonymous by default.